This study seeks to develop and test through extensive simulations a method of finding accurate, time-varying, inverse solutions to the electrocardiographic problem. The goals are to minimize inhomogeneity effects, the consequences of variable torso shapes, and the uncertainties associated with the natural ambiguities of the electrocardiographic problem. Additional goals are those of validating a number of electrocardiographic models. The UVM inhomogenous anisotropic torso model data will be utilized as a standard of comparison against which computer calculations of transfer coefficients will be evaluated.